An early warning system for detecting leakage in foundation pits

Abstract

This invention proposes an early warning system for foundation pit leakage detection, addressing the technical problems of low reliability, low monitoring efficiency, and difficulty for workers to locate leakage points in automated foundation pit detection. The invention includes a data acquisition unit, a processing unit, and an early warning unit. Based on the acquired information, it obtains state evaluation values ​​for each location on the foundation pit sidewall and bottom. Leakage points are located based on these state evaluation values. The seepage path is then derived from the leakage point as the starting point. The seepage path is substituted into a calculation function to obtain evaluation coefficients. By setting weights and combining them with the evaluation coefficients, an overall evaluation coefficient is obtained. The dispersion of the overall evaluation coefficients is used to find the central path and center point of the seepage. Finally, a piecewise function is plotted using a prediction network to predict the future state trends of the foundation pit sidewall and bottom, providing early warning. This invention achieves high reliability and high monitoring efficiency in automated foundation pit detection, while also facilitating the location of leakage points for workers.

Description

Technical Field

[0001] This invention relates to the technical field of foundation pit leakage detection, and in particular to an early warning system for foundation pit leakage detection. Background Technology

[0002] During foundation pit excavation, water seepage can alter the groundwater level in the vicinity of buildings, causing a drop in the water level and resulting in soil consolidation and settlement. Simultaneously, water seepage outside the foundation pit can also cause soil loss. If the excavation is too deep and lacks timely protection from support structures, the passive earth pressure within the pit may be insufficient to resist the active earth pressure generated by nearby buildings and the soil beneath them, leading to displacement of the soil above the pit. Both of these phenomena can result in uneven settlement of the building's foundation soil, causing the building to tilt, crack, or collapse. Therefore, detecting foundation pit seepage is of paramount importance.

[0003] For example, Chinese invention patent CN111535376B, with an authorization announcement date of September 14, 2021, discloses an automated monitoring and control method for foundation pits and its construction method. The method includes a foundation pit, a central calibration and comparison device located in the middle of the bottom wall of the foundation pit, sampling and fixed-point detection devices symmetrically located on both sides of the bottom wall of the foundation pit, and several evenly distributed side wall data detection devices located inside the side walls of the foundation pit. A mounting frame is provided on one side of each side wall data detection device, and an infrared comparison calibration box is provided inside the mounting frame to cooperate with the sampling and fixed-point detection device. A signal receiver is provided at the top of the infrared comparison calibration box, and a signal transmitter is provided at the bottom of the infrared comparison calibration box.

[0004] Although the aforementioned invention patent can also automatically monitor the current usage status of the pit sidewalls and bottom, it ignores the impact of extreme external weather on the automated monitoring of the pit and does not take into account the external environment of the pit. This greatly reduces the reliability of the invention patent for automated monitoring of the pit. At the same time, when a leakage area appears, the specific leakage point cannot be found immediately, resulting in reduced monitoring efficiency and increased difficulty for workers when sealing the leak. Summary of the Invention

[0005] In view of the shortcomings in the background technology, the present invention proposes an early warning system for foundation pit leakage detection, which solves the technical problems of low reliability, low monitoring efficiency and difficulty for staff to find leakage points in automated foundation pit monitoring.

[0006] To achieve the above objectives, the present invention provides an early warning system for detecting leakage in foundation pits, comprising:

[0007] Data Acquisition Unit: Sensors are arranged on the sidewalls and bottom of the foundation pit to acquire data information at each location on the sidewalls and bottom of the foundation pit. The sensors include humidity sensors for acquiring humidity information at each location on the sidewalls and bottom of the foundation pit, and resistance testers for acquiring soil resistivity information at each location on the sidewalls and bottom of the foundation pit. A water level sensing float is placed in the groundwater in the foundation pit observation well. The water level sensing float is used to acquire water level height information. A capacitive sensor is placed inside the water level sensing float to acquire groundwater flow information.

[0008] Processing Unit: Collects and preprocesses the moisture, soil resistivity, water level, and groundwater flow information of the pit sidewalls and bottom collected by the acquisition unit. This preprocessing yields the following data sequences: moisture, soil resistivity, water level, and groundwater flow. Based on these sequences, a state evaluation value sequence is obtained for each location on the pit sidewalls and bottom. A groundwater state evaluation value sequence is also obtained based on the water level and groundwater flow data sequences.

[0009] Construct a computational model: Collect the information obtained from preprocessing in the processing unit, obtain the ratio of the state evaluation values ​​of the pit sidewall and pit bottom at adjacent locations, find the leakage point based on the ratio, and divide all locations into at least three clusters based on the leakage point and the state evaluation value of each location of the pit sidewall and pit bottom.

[0010] A ray is constructed starting from the leakage point, and the output values ​​of the state evaluation data of each position on the sidewall and bottom of the pit along the ray are obtained. The seepage boundary line is obtained based on the obtained output values, and the seepage path is obtained according to the leakage point and the seepage boundary line.

[0011] Substituting the seepage path into the calculation function yields the optimal solution function model. Differentiating the function model yields the evaluation coefficient for each location along the seepage path. The groundwater state evaluation value is then set as the weight. This weight is compared with the evaluation coefficient for each location. An abnormal weight indicates increased humidity and soil resistivity at each location on the pit sidewall and bottom due to rainfall, causing abnormal evaluation coefficients for each location along the seepage path, which does not need to be considered. The evaluation coefficient for each location along each seepage path is obtained. Based on the standard deviation of the evaluation coefficients for all locations along each seepage path, the overall evaluation coefficient for each seepage path is obtained.

[0012] The dispersion of each seepage path is calculated based on the overall evaluation coefficient of each seepage path, and the central path and center point of the seepage are found by the dispersion.

[0013] Early warning unit: Receives information from the computational model and assigns evaluation coefficients for each location along the seepage path.

[0014] The evaluation coefficients of each location on the future seepage path are obtained by predicting the network. Piecewise functions are plotted to obtain the trend of the state evaluation values ​​of each location on the future seepage path and the bottom of the pit. Early warnings are given based on the state of the pit sidewalls and bottom in the future time period.

[0015] Furthermore, the method for obtaining the leakage point is as follows:

[0016] If the ratio of the condition evaluation values ​​of the current location's pit sidewall and bottom to those of its adjacent locations is greater than 1, then the current location is a seepage point. This indicates that the condition evaluation values ​​of the current location's pit sidewall and bottom are greater than the condition evaluation values ​​of the four surrounding locations, which is consistent with the characteristic of groundwater seeping into the seepage point and gradually permeating outwards. Therefore, the current location is recorded as the seepage point of groundwater infiltration. The adjacent locations include the four locations adjacent to the current location: east, south, west, and north.

[0017] Furthermore, the step of obtaining clusters includes:

[0018] Based on the difference in state evaluation values ​​between every two locations on the pit sidewall and the pit bottom as the distance, a minimum of three clusters are obtained by using the DBSCAN clustering algorithm with a preset search radius.

[0019] Furthermore, the method for obtaining the function model of the optimal solution is as follows:

[0020]

[0021] Where: i represents the location number corresponding to the state evaluation value of each foundation pit sidewall and pit bottom; U i This represents the state evaluation value of the pit sidewall and pit bottom corresponding to the i-th location label.

[0022] Furthermore, the method for combining the groundwater state evaluation value as a weight with the evaluation coefficient obtained at each location is as follows:

[0023] The state evaluation values ​​of groundwater over a period of time are obtained, resulting in a sequence of groundwater state evaluation values. Based on the variance of the groundwater state evaluation value sequence, the stability of the sequence over a period of time is determined. A threshold is set and compared with the stability of the groundwater state evaluation value sequence over a period of time. Values ​​exceeding the threshold are considered weighted anomalies.

[0024] Furthermore, the method for obtaining the piecewise function is as follows:

[0025]

[0026] Among them: U 未 The values ​​are the condition evaluation values ​​for each location on the future foundation pit sidewall and bottom, with 0.5β, 1β, and 2β being variation coefficients, and X being the evaluation coefficient for each location on the future seepage path.

[0027] Furthermore, the method for obtaining the degree of dispersion is as follows:

[0028]

[0029] Where: F is the overall evaluation coefficient of each seepage path, and K is the dispersion of each seepage path.

[0030] The present invention has at least the following beneficial effects: It not only collects and analyzes the condition of the pit sidewalls and bottom, but also considers the weather conditions surrounding the pit, providing a multi-level and multi-angle analysis of the pit's condition, greatly improving the reliability and accuracy of automated pit detection results; furthermore, by analyzing the dispersion of the overall evaluation coefficients, it identifies the central path and center point of seepage, resulting in high monitoring efficiency and significantly reducing the difficulty for workers to repair leaks; through a prediction network, it obtains the future trend of the condition evaluation values ​​for each location on the pit sidewalls and bottom based on the evaluation coefficients at each location along the future seepage path, plotting a piecewise function that is intuitive and clear, facilitating pit condition analysis for workers, providing early warnings, and reducing losses. Attached Figure Description

[0031] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0032] Figure 1 A system schematic diagram provided for this invention. Detailed Implementation

[0033] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0034] like Figure 1 As shown, an early warning system for detecting leakage in foundation pits includes:

[0035] Data Acquisition Unit: Sensors are arranged on the sidewalls and bottom of the foundation pit to acquire data information at each location on the sidewalls and bottom of the foundation pit. The sensors include humidity sensors for acquiring humidity information at each location on the sidewalls and bottom of the foundation pit, and resistance testers for acquiring soil resistivity information at each location on the sidewalls and bottom of the foundation pit. A water level sensing float is placed in the groundwater in the foundation pit observation well. The water level sensing float is used to acquire water level height information. A capacitive sensor is placed inside the water level sensing float to acquire groundwater flow information.

[0036] Processing Unit: Collects and preprocesses the moisture, soil resistivity, water level, and groundwater flow information of the pit sidewalls and bottom collected by the acquisition unit. This preprocessing yields the following data sequences: moisture, soil resistivity, water level, and groundwater flow. Based on these sequences, a state evaluation value sequence is obtained for each location on the pit sidewalls and bottom. A groundwater state evaluation value sequence is also obtained based on the water level and groundwater flow data sequences.

[0037] Construct a computational model: Collect the information obtained from preprocessing in the processing unit, obtain the ratio of the state evaluation values ​​of the pit sidewall and pit bottom at adjacent locations, find the leakage point based on the ratio, and divide all locations into at least three clusters based on the leakage point and the state evaluation value of each location of the pit sidewall and pit bottom.

[0038] A ray is constructed starting from the leakage point, and the output values ​​of the state evaluation data of each position on the sidewall and bottom of the pit along the ray are obtained. The seepage boundary line is obtained based on the obtained output values, and the seepage path is obtained according to the leakage point and the seepage boundary line.

[0039] Substituting the seepage path into the calculation function yields the optimal solution function model. Differentiating the function model yields the evaluation coefficient for each location along the seepage path. The groundwater state evaluation value is then set as the weight. This weight is compared with the evaluation coefficient for each location. An abnormal weight indicates increased humidity and soil resistivity at each location on the pit sidewall and bottom due to rainfall, causing abnormal evaluation coefficients for each location along the seepage path, which does not need to be considered. The evaluation coefficient for each location along each seepage path is obtained. Based on the standard deviation of the evaluation coefficients for all locations along each seepage path, the overall evaluation coefficient for each seepage path is obtained.

[0040] The dispersion of each seepage path is calculated based on the overall evaluation coefficient of each seepage path, and the central path and center point of the seepage are found by the dispersion.

[0041] Early warning unit: Obtains information from the computational model, uses the evaluation coefficient of each location on the seepage path to obtain the evaluation coefficient of each location on the future seepage path through the prediction network, plots a piecewise function to obtain the trend of the future state evaluation value of each location of the foundation pit sidewall and bottom based on the obtained evaluation coefficient of each location on the future seepage path, and provides early warning based on the state of the foundation pit sidewall and bottom in the future time period.

[0042] Furthermore, the processing unit adopts TPS22976DPUR. The TPS22976DPUR integrated data processing unit has low cost, small size and convenient use. In this invention, the Internet of Things is used to transmit and display the information processed by the processing unit. The data processing in this invention is not complicated, so a high-performance processing unit is used to further reduce costs.

[0043] Furthermore, the sidewalls and bottom of the foundation pit are evenly divided into several areas, and each area is equipped with a humidity sensor and a soil resistivity grounding electrode to form a monitoring network to monitor each area of ​​the foundation pit sidewalls and bottom.

[0044] Furthermore, before leakage occurs in the foundation pit, the humidity in various areas of the pit's sidewalls and bottom will become abnormal. Therefore, humidity sensors are used to detect whether there is leakage in the pit's sidewalls and bottom. When the humidity sensor in the foundation pit malfunctions, it indicates that there is water seepage in the foundation pit.

[0045] Furthermore, soil resistivity is one of the fundamental properties of soil, representing the resistance per unit cubic meter of soil. Many factors influence soil resistivity, such as soil moisture content, soil temperature, and soil dielectric constant. When leakage occurs within the foundation pit, the soil moisture content and dielectric constant will become abnormal. Therefore, collecting soil resistivity data is crucial to reflecting the presence of leakage within the foundation pit. When measuring soil resistivity using a resistance tester, a known grounding electrode for the tester is buried at the location where soil resistivity measurement is needed. In a sensor-based Internet of Things (IoT) system, the resistance tester is connected to a computer using a TKM-100 wireless data transmission module, feeding the collected information back to the computer analysis terminal.

[0046] Furthermore, this embodiment combines soil moisture information with soil resistivity information. This is mainly to avoid the influence of using only one type of information when different geological conditions or soil types occur. When the soil is relatively loose, with a certain porosity, the lower the moisture content, the higher the dielectric constant. Both the dielectric constant and the volumetric moisture content of the soil are factors affecting soil resistivity. Therefore, when the soil type is different, the volumetric moisture content and soil resistivity are not linearly related. Different soils have different internal water contents when moisture saturation occurs. Therefore, this embodiment combines soil moisture information with soil resistivity information to jointly verify whether there is leakage inside the foundation pit. This multi-level and multi-angle analysis makes the analysis results more comprehensive and improves the authenticity of the analysis results.

[0047] Furthermore, groundwater below the excavation pit is monitored using a water level sensing float. The float is deployed into the groundwater through an observation well in the pit, and a capacitive sensor is placed inside to monitor groundwater flow. When it rains, the groundwater level rises and the flow rate increases. This not only increases the groundwater status evaluation value, but also increases the soil resistivity and humidity data at each location on the pit's sidewalls and bottom. The evaluation coefficients at each location along the seepage path, derived from the soil resistivity and humidity data, will also show anomalies, causing the system to detect seepage and issue an early warning to staff. The groundwater status evaluation value obtained from the monitoring data is used as a weight. Anomalies in the weights indicate that the increased humidity and soil resistivity at each location on the pit's sidewalls and bottom are due to rain, causing the abnormal evaluation coefficients at each location along the seepage path, rather than seepage from within the pit itself. In this invention, the Internet of Things (IoT) is used to transmit the collected information to the processing unit for processing.

[0048] Furthermore, the method for obtaining the leakage point is as follows:

[0049] If the ratio of the condition evaluation values ​​of the current location's pit sidewall and bottom to those of its adjacent locations is greater than 1, then the current location is a leakage point. This indicates that the condition evaluation values ​​of the current location's pit sidewall and bottom are greater than those of the four surrounding locations, consistent with the characteristic of groundwater seeping into the leakage point and gradually infiltrating outwards. Therefore, the current location is recorded as the groundwater seepage point. Adjacent locations include the four cardinal directions (east, south, west, and north) adjacent to the current location. A higher condition evaluation value for each location on the pit sidewall and bottom indicates more severe leakage within the pit.

[0050] Furthermore, when leakage occurs on the sidewalls and bottom of the foundation pit, the water in the confined aquifer has a high permeability. During the infiltration process, the water in the confined aquifer gradually seeps from the seepage point to the surface. Therefore, the groundwater infiltration leakage point can be obtained based on the state evaluation value of each location on the sidewalls and bottom of the foundation pit. After obtaining the leakage point, the analysis method of connecting the points to the surface is adopted for the sidewalls and bottom of the foundation pit, which makes the analysis more three-dimensional.

[0051] Furthermore, the step of obtaining clusters includes: using the difference in state evaluation values ​​between every two locations on the pit sidewall and the pit bottom as the distance, and using the DBSCAN clustering algorithm with a preset search radius to obtain at least three clusters.

[0052] Furthermore, DBSCAN is used to find at least three clusters with a preset radius. The specific number of clusters can be adjusted by the implementer according to different site conditions. In this embodiment, three clusters are preset to be found. Clustering the data can effectively help classify the data, reduce the calculation time, and improve the calculation efficiency.

[0053] Furthermore, the method for obtaining the function model of the optimal solution is as follows:

[0054]

[0055] Where: i represents the location number corresponding to the state evaluation value of each foundation pit sidewall and pit bottom; U i This represents the state evaluation value of the pit sidewall and pit bottom corresponding to the i-th location label.

[0056] Furthermore, the method for combining the groundwater state evaluation value as a weight with the evaluation coefficient obtained at each location is as follows:

[0057] The state evaluation values ​​of groundwater over a period of time are obtained, resulting in a sequence of groundwater state evaluation values. Based on the variance of the groundwater state evaluation value sequence, the stability of the sequence over a period of time is determined. A threshold is set and compared with the stability of the groundwater state evaluation value sequence over a period of time. Values ​​exceeding the threshold are considered weighted anomalies.

[0058] Furthermore, the method for obtaining the piecewise function is as follows:

[0059]

[0060] Among them: U 未 The values ​​are the condition evaluation values ​​for each location on the future foundation pit sidewall and bottom, with 0.5β, 1β, and 2β being variation coefficients, and X being the evaluation coefficient for each location on the future seepage path.

[0061] Furthermore, after obtaining the evaluation coefficients for each location on the future seepage path through the prediction network, the state evaluation values ​​for each location on the future foundation pit sidewall and bottom are obtained through the change coefficients, and a piecewise function is plotted, which is clear and intuitive.

[0062] Furthermore, after obtaining the evaluation coefficients for each location on the future seepage path through the prediction network, the state evaluation values ​​for each location on the future foundation pit sidewall and bottom are obtained through the change coefficients, and a piecewise function is plotted, which is clear and intuitive.

[0063] Furthermore, the piecewise function represents the first stage as the seepage point, the second stage as the seepage hole, and the third stage as the seepage surface. When a seepage point appears, if grouting is not performed to seal it in time, the seepage will gradually spread from the point to the surface. At this time, the water inside the foundation pit will seep into the soil roots, causing a large-scale collapse and affecting the safety of the entire foundation pit. The result is uneven settlement of the foundation soil, which in turn causes the building to tilt, crack, or collapse.

[0064] Furthermore, the method for obtaining the degree of dispersion is as follows:

[0065]

[0066] Where: F is the overall evaluation coefficient of each seepage path, and K is the dispersion of each seepage path.

[0067] Furthermore, the overall evaluation coefficient of each seepage path is obtained. All evaluation coefficients are used to calculate the sum of the distances between the current seepage path and other seepage paths using a formula. Then, the sum of the distances from all seepage paths to other seepage paths is analyzed, and the path with the smallest sum of distances to other paths is identified as the seepage center path. The leakage point on the seepage path is the seepage center point.

[0068] Furthermore, in this embodiment, the sidewalls and bottom of the foundation pit are evenly divided into 1-square-meter unit areas. The specific division of unit areas can be adjusted according to different implementers. The grounding body of the humidity sensor and the resistance tester is set in the middle of the unit area. The humidity and soil resistivity data are collected through the grounding body of the humidity sensor and the resistance tester. At the same time, the water level sensing float is deployed into the groundwater through the foundation pit observation well. The displacement and vibration information of the water level sensing float are collected by the internal capacitive sensor. In this invention, the Internet of Things is used to transmit the collected information to the processing unit for processing. After analysis, the processing unit determines whether there is any abnormality in the groundwater at this time, and then determines whether the evaluation coefficient of each location is accurate.

[0069] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A pre-warning system for detecting leakage of a foundation pit, characterized in that, The early warning system comprises: The acquisition unit: sensors are arranged on the side wall and bottom of the foundation pit to obtain data information of each position of the side wall and bottom of the foundation pit, the sensors adopt humidity sensors for collecting humidity information of each position of the side wall and bottom of the foundation pit, and grounding bodies of resistance testers for collecting soil resistivity information of each position of the side wall and bottom of the foundation pit; a water level sensing float ball is put into underground water in a foundation pit observation well, the water level sensing float ball is used for collecting water level height information, and a capacitance sensor is placed in the water level sensing float ball, which is used for collecting underground water flow information; The processing unit: the humidity information of the side wall and bottom of the foundation pit, the soil resistivity information of the side wall and bottom of the foundation pit, the water level height information and the underground water flow information collected by the acquisition unit are collected for preprocessing, and the humidity data sequence of the side wall and bottom of the foundation pit, the soil resistivity data sequence of the side wall and bottom of the foundation pit, the water level height data sequence and the underground water flow data sequence are obtained after preprocessing, the state evaluation value sequence of each position of the side wall and bottom of the foundation pit is obtained according to the humidity data sequence of the side wall and bottom of the foundation pit and the soil resistivity data sequence of the side wall and bottom of the foundation pit, and the state evaluation value sequence of underground water is obtained according to the water level height data sequence and the underground water flow data sequence; The calculation model is constructed: the information obtained by preprocessing in the processing unit is collected, the ratio of the state evaluation values of the side wall and bottom of the foundation pit of adjacent positions is obtained, the leakage point is obtained according to the ratio, and all positions are divided into at least three cluster clusters according to the leakage point and the state evaluation values of each position of the side wall and bottom of the foundation pit; The leakage point is taken as a starting point to construct a ray, the output value of the state evaluation value data of each position of the side wall and bottom of the foundation pit on the ray is obtained, the water seepage boundary line is obtained based on the obtained output value, and the water seepage path is obtained according to the leakage point and the water seepage boundary line; The water seepage path is substituted into a function model of an optimal solution to obtain the evaluation coefficient of each position on the water seepage path by derivation, the state evaluation value of underground water is taken as a weight at this time, and the evaluation coefficient of each position is obtained at this time to make a judgment, the weight appears abnormal, which proves that the humidity and soil resistivity of each position of the side wall and bottom of the foundation pit increase due to rain, causing the evaluation coefficient of each position on the water seepage path to be abnormal and not needing to be considered; the evaluation coefficient of each position on each water seepage path is obtained, and the overall evaluation coefficient of each water seepage path is obtained according to the standard deviation of the evaluation coefficients of all positions on each water seepage path; The dispersion degree of each water seepage path is calculated according to the overall evaluation coefficient of each water seepage path, and the central path and central point of water seepage are found through the dispersion degree; The early warning unit: information fed back by the calculation model is obtained, the evaluation coefficient of each position on the water seepage path is obtained through a prediction network to obtain the evaluation coefficient of each position on the future water seepage path, a segmented function is drawn to obtain the trend of the state evaluation value of each position of the side wall and bottom of the foundation pit in the future according to the evaluation coefficient of each position on the future water seepage path, and early warning is made according to the state of the side wall and bottom of the foundation pit in the future time period.

2. The early warning system for foundation pit leakage detection according to claim 1, characterized in that, The method for obtaining the leakage point is: If the ratio of the state evaluation value of the current position of the foundation pit side wall and pit bottom to the state evaluation value of its adjacent position of the foundation pit side wall and pit bottom is greater than 1, the current position is a leakage point, indicating that the state evaluation value of the current position of the foundation pit side wall and pit bottom is greater than that of the surrounding four positions, which is consistent with the characteristics of groundwater infiltration into the leakage point and gradually penetrating in all directions. The current position is recorded as the groundwater infiltration leakage point. The adjacent positions include the southeast, southwest and northwest positions adjacent to the current position.

3. The early warning system for foundation pit leakage detection according to claim 1, characterized in that, The step of obtaining the clustering cluster comprises: Based on the difference between the state evaluation values of every two positions of the foundation pit side wall and pit bottom as the distance, the preset search radius is obtained by the DBSCAN clustering algorithm to obtain at least three clustering clusters.

4. The early warning system for foundation pit leakage detection according to any one of claims 1-3, characterized in that, The method for obtaining the function model of the optimal solution is: Wherein: i represents the position mark corresponding to the state evaluation value of each foundation pit side wall and pit bottom; U i represents the state evaluation value of the foundation pit side wall and pit bottom corresponding to the i-th position mark.

5. The early warning system for foundation pit leakage detection according to claim 4, characterized in that, The state evaluation value of the groundwater is set as the weight, and the method for combining the evaluation coefficient of each position obtained at this time is: The state evaluation value of the groundwater in a period of time is obtained to obtain the state evaluation value sequence of the groundwater. According to the variance of the state evaluation value sequence of the groundwater, the stability degree of the sequence in a period of time is obtained. The threshold value is set to compare the stability degree of the state evaluation value sequence of the groundwater in a period of time. If the threshold value is exceeded, it is set as the weight anomaly.

6. The early warning system for foundation pit leakage detection according to claim 5, characterized in that, The method for obtaining the piecewise function is: Wherein: U 未 is the state evaluation value of each position of the future foundation pit side wall and pit bottom, 0.5β, 1β, 2β are change coefficients, and X is the evaluation coefficient of each position on the future water seepage path.

7. The early warning system for foundation pit leakage detection according to claim 1, characterized in that, The method for obtaining the discrete degree is: Wherein: F is the overall evaluation coefficient of each water infiltration path, and K is the discrete degree of each water infiltration path.

Images